English(EN)Same Model, Different Weakness: How Language and Modality Reshape the Jailbreak Attack Surface in Frontier MLLMs
新研究探索用于大语言模型（LLM）越狱检测和缓解的先进方法
作者PulseAugur 编辑部·[8 个来源]·
研究人员正在开发检测和缓解针对大语言模型（LLMs）的越狱攻击的新方法。一种名为SelfGrader的方法使用锚定令牌级对数概率来评估查询安全性，具有低延迟和低开销。另一项研究探讨了多模态大语言模型（MLLMs）的不同设计范式，特别是显式的图像-工具交互，如何提高对抗越狱的鲁棒性。此外，还提出了一个名为“行为几何”的框架，用于在模型群体之间进行有效的易感性预测和防御迁移。最后，研究表明语言和模态相互作用，共同塑造了多模态大语言模型（MLLMs）的攻击面，这表明安全评估需要跨语言进行并考虑这些相互作用。
AI
arXiv:2604.01473v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are powerful tools for answering user queries, yet they remain highly vulnerable to jailbreak attacks. Existing guardrail methods typically rely on internal features or textual responses to det…
arXiv:2605.27932v1 Announce Type: cross Abstract: Think-with-image reasoning is emerging as a new inference paradigm for large vision-language models, but its safety implications remain poorly understood. Existing systems already span multiple process designs, including direct re…
arXiv:2605.26409v1 Announce Type: cross Abstract: Evaluating and mitigating a generative system's susceptibility to jailbreak attacks is critical to its safe deployment. Given the number of deployable systems, full per-configuration evaluation and optimization is impractical. In …
arXiv:2605.10764v2 Announce Type: replace-cross Abstract: Recent studies show that gradient-based universal image jailbreaks on vision-language models (VLMs) exhibit little or no cross-model transferability, casting doubt on the feasibility of transferable multimodal jailbreaks. …
arXiv cs.AI
TIER_1English(EN)·Seokil Ham, Jaehyuk Jang, Wonjun Lee, Changick Kim·
arXiv:2605.24550v1 Announce Type: new Abstract: Fine-tuning-as-a-Service (FaaS) enables personalization of large language models (LLMs), but it can weaken safety-alignment under harmful fine-tuning attacks. Recent work has shown that activating harmful-behavior modules during fin…
arXiv:2506.18543v2 Announce Type: replace-cross Abstract: The rapid proliferation of Large Language Models (LLMs) has heightened concerns regarding their exposure to jailbreak attacks, which craft adversarial inputs designed to elicit unsafe content. Although proprietary models s…
arXiv cs.CL
TIER_1English(EN)·Casey Ford, Madison Van Doren, Sicheng Jin, Emily Dix·
arXiv:2605.23157v1 Announce Type: new Abstract: The attack surface of a multimodal large language model (MLLM) is language-dependent in ways that reveal the mechanistic structure of alignment failures. We present the first systematic cross-lingual, multimodal red-teaming study co…
The attack surface of a multimodal large language model (MLLM) is language-dependent in ways that reveal the mechanistic structure of alignment failures. We present the first systematic cross-lingual, multimodal red-teaming study comparing jailbreak vulnerability in US English (e…
